The continuing expansion of the range of applications for computers, information technology, and entertainment has created a growing need for faster and more efficient storage and retrieval systems for processing information and data. Of paramount importance to these systems are recording media that are capable of storing as much information as possible while still permitting the quick recording, reading, and erasing of information. Optical recording media offer the most promise for future information storage needs. They currently allow for reading, writing, and erasing of information through all-optical means.
In order for an optical recording medium to be successful, it must be possible to read, write, and erase information quickly and with high fidelity over an extended number of storage and retrieval cycles. This performance capability requires an optical recording medium that can reproducibly and reversibly undergo a physical or chemical transformation in the presence of light. Phase change materials are currently emerging as the leading rewritable optical recording medium. They function by reversibly transforming between two physical states under the influence of a modulated light beam. In the most common phase change materials, the transformation occurs between an amorphous phase and a crystalline or partially crystalline phase. Under typical operating conditions, the amorphous phase corresponds to the recorded state and the crystalline or partially crystalline phase corresponds to the erased state. The process of writing or recording typically involves exposing a portion of the phase change material to a laser beam with sufficient energy to heat the portion of material above its melting point. Subsequent removal of the laser leads to rapid quenching and formation of an amorphous state characterized by structural randomness and disorder. The amorphous state is structurally distinct from the surrounding portions of the phase change material and provides a region of contrast that can be distinguished by a property such as reflectivity. The amorphous regions created by the write beam are frequently referred to as amorphous marks and may be viewed as corresponding to regions of stored information on the recording medium. By applying the write beam to selected portions of the recording medium, a pattern of amorphous marks among crystalline or partially crystalline spaces may be formed that corresponds to the particular information that one wishes to store. Properties such as the length, width, and spacing between amorphous marks may be used to encode information.
The process of reading requires identification of the pattern of amorphous marks and crystalline spaces present on the phase change material. This process entails detecting the contrast between the amorphous marks and the surrounding portions of the phase change material. Detection involves recognizing a difference in at least one property of the amorphous marks relative to the surrounding portions of the material. Differences in state are detected by differences in properties such as electrical resistivity, optical reflectivity or optical transmissivity. Detection of optical properties by an optical read beam is preferred because it permits construction of all-optical storage and retrieval systems. When a laser is used as the read beam, its power is set sufficiently low to avoid changing the physical state of the phase change material. The read beam power, for example, must be low enough to prevent melting and inadvertent formation of unintended amorphous marks or the formation of crystalline regions within amorphous marks.
The process of erasing requires removal of amorphous marks and typically involves transforming the recorded amorphous phase of the phase change material into a crystalline or partially crystalline phase. The transformation can be accomplished optically, for example, by an erase beam provided by a laser that has enough power to heat the amorphous phase above its crystallization temperature, but below its melting temperature. Heating of the amorphous marks with an erase beam to a temperature between the crystallization and melting temperatures provides enough energy to promote the atomic motion necessary for structural reorganization to a crystalline or partially crystalline phase without creating a high mobility or melt state that is susceptible to reforming an amorphous phase upon removal of the erase beam. The crystalline or partially crystalline erased regions may be understood as corresponding to unrecorded regions of the phase change material and are distinguishable from amorphous marks by at least one physical property such as electrical resistivity, optical reflectivity, or optical transmissivity. In an all-optical system, use of a read beam to spatially probe a property such as optical reflectivity or optical transmissivity permits the recognition and distinguishing of the recorded and unrecorded portions of the phase change material. Erasure is not the only process for removing an existing pattern of amorphous marks. Alternatively, an existing pattern can simply be directly overwritten by a new pattern representing new information.
Several chalcogenide-based materials have been demonstrated to function effectively as phase change materials. These materials are capable of existing in the amorphous, crystalline and partially crystalline states at room temperature and are readily transformable between the amorphous phase and the crystalline or partially crystalline phase. Representative phase change materials include alloys containing one or more of the elements Ge, Te, Sb, Se, S, Bi, In, Ga, Ag, Si, and As.
CD technology was the first widely used optical recording technology and is currently being replaced with newer high data density DVD technology. Although the currently available optical phase change materials provide excellent read, write and erase characteristics, further improvements are possible and desirable for future information storage needs. One area of possible improvement is information storage density. The currently available optical phase change materials may provide much higher information storage densities than many conventional magnetic memory media, but are currently limited. The storage density, for example, may be limited by the wavelength of light used to record or write information to a phase change material. The longer the wavelength of light, the larger the size of the recorded amorphous mark and the lower the information storage density. The wavelength of light used in current phase change materials is controlled largely by the availability of semiconductor diode lasers with sufficient power to effect the necessary structural transformations between the recorded and unrecorded states. The most suitable lasers currently available operate in the red-near infrared portion of the visible spectrum. The writing of CDs, for example, may be done with lasers that operate at about 780 nm or 830 nm. This wavelength of light provides recorded marks on a length scale of about 1 micron. More recent DVDs, in contrast, use 650 nm light and contain recorded marks on a length scale of about 0.5 micron. By decreasing the wavelength of light, it becomes possible to further decrease the size of recorded amorphous marks and increase the information storage density.
Achieving shorter write beam wavelengths requires the development of both economical compact short wavelength lasers and new phase change materials capable of functioning at shorter wavelengths. Recent developments in the field of semiconductor lasers indicate that compact lasers based on GaN that operate in the blue region of the visible spectrum may be useful in reading data from high density optical media. Unfortunately, high power blue lasers are not currently economically feasible for use in mass produced products. Additionally, present day phase change materials have lower contrast than is preferred to adequately differentiate between the initialized and written or written and erased states at such shorter wavelengths. Consequently, it is desirable to develop new phase change materials and optical storage and retrieval systems that can fully realize the much higher information storage densities potentially available from blue laser sources.
Another method of increasing the is to decrease the spot size of present light sources. In any recording technology, the physical dimension of the minimal mark size ultimately governs the storage density. Thus in current generation DVD products, the minimal mark of about 400 nm diameter can achieve about 4.7 GB per side of a 12 cm diameter disk, representing about 2.7 Gbits/in2 for binary recording. In multi-level (ML) recording, a range of mark sizes are produced between 400 nm diameter and a minimum mark of about 150 nm. This results in an increase of about 2× in the storage capacity over binary recording. The capacity is not greater than this in ML recording because each mark must still be resolved by the optics, meaning that marks of any size must be spaced sufficiently apart. In practice, they are spaced on a constant “data cell” dimension of about 400 nm, i.e., equal to the minimal mark size in binary recording.
In the DVD products and the ML enhancement, the resolution is governed by diffraction limited optics using 650 nm wavelength light, and an objective lens having 0.6 numerical aperture (NA). Strictly speaking, with these optical parameters, the Gaussian input beam can be focused to a full width at half maximum of about 540 nm (=λ/2NA). Controlling the power and temporal character of the applied Gaussian beam profile can result in significantly smaller marks than what can be resolved by the optics. But the spacing of the ML marks and the minimal mark size in binary recording must both substantially conform to the resolution limit of the optics. Clearly the 400 nm dimension is pushing this limit.
Recent activity to push the optical storage density even higher is based on at least three different strategies: (1) pushing standard “far-field” optics; (2) developing “standard near-field” optics; and (3) developing aperture optics, which is also a near-field concept. To push standard “far-field” optics, the light's wavelength and the lens's NA are critical components to greater resolution. Reducing the wavelength to about 400 nm, and increasing the NA to 0.85, results in a 2.3× improvement in resolution for a storage capacity of about 25 GB per side of 12 cm diameter disk (˜14 Gbits/in2). In developing “standard near-field” optics the NA's are pushed beyond 1.0 using a solid immersion lens (sil) to approach a capacity of about 35 GB using the blue laser light. Here, evanescent coupling between the sil and the active phase change layer requires very short distances between the lens and the media of less than 100 nm. In one novel approach, the near field optic is incorporated into the disk itself, thereby eliminating the need for such fine control of the optics.
Finally in developing aperture optics, the mark dimension is not governed by optical principles of lenses, but by apertures. Here, in “Super-RENS” technology, the disk contains one low melting temperature layer such as Sb that acts as a controllable aperture. By use of the Gaussian profile laser beam, small apertures can be obtained with resolution limits of less than 100 nm using 635 nm wavelength light. (See J. Tominaga et al., Appl. Phys. Lett. 73, 2078 (1998)). Tominaga et al., have recently proposed to achieve greater read back signal strength by further enhancing the optical coupling using surface plasmons. In their work, geometric constraints do not allow surface plasmons (SP) on the metallic Sb layer because momentum is not conserved. Instead, they propose to use marks recorded in the GeSbTe phase change layer itself to act as a grating in which to generate the SP's. The SP's then localize preferentially around smaller marks, thereby amplifying their effective coupling cross section. It is not clear how effective using the marks themselves will actually be to generate the grating. This requires a change in sign of the real part of the dielectric function between the two relevant components (i.e., between crystalline and amorphous states of the GeSbTe compound). The optical constants used in their analysis correspond to the hexagonal and amorphous states. They did not use the crystalline state of the GeSbTe material that is relevant in the phase change applications, namely the fcc state. The real part of the dielectric function is positive for both the fcc and amorphous states, making generation of SP's unlikely.
A separate developmental front has proceeded with surface plasmons themselves that ultimately may play a role in optical memory applications. Recently Ebbesen et al. showed that normally incident light can transmit (up to 10% level) through a flat optically thick metal film that is perforated with many small holes, even though their diameters are roughly a factor of 10 less than the light's wavelength (see T. W. Ebbesen et al., “Extraordinary optical transmission through sub-wavelength hole arrays”, Nature 391, 667 (1998) and H. F. Ghaemi et al, “Surface plasmons enhance optical transmission through subwavelength holes”, Phys. Rev. B 58, 6779(1998)). These diameters are significantly smaller than the cut-off dimension that would allow propagation. Usually, light incident normally to a metal surface cannot couple with surface plasmons, which are longitudinal waves along the surface where the electric field and propagation vectors are parallel, because light propagates as transverse waves whose electric and propagation vectors are perpendicular. Another way to say this is that coupling does not occur because energy and momentum are not simultaneously conserved. By scattering off the holes in a grating however, the light's momentum can gain some component parallel to the metal surface through the grating “momentum”.
Once it became clear the importance of the grating momentum in coupling with SP's, it was realized that the scattering sites do not need to be holes through the metal film. In fact, any periodic (i.e. diffraction-like) grating can stimulate surface plasmons. Indeed, the NEC group has shown how a 2-D array of pits can be used to enhance the transmission of light through a single hole. See D. E. Grupp et al., “Beyond the Bethe Limit: Tunable Enhanced Light Transmission Through a Single Subwavelength Aperture”, Adv. Mater. 11, 860 (1999); Thio et al., “Strongly Enhanced Optical Transmission Through Subwavelength Holes in Metal Films”, Physica B, 279, 90 (2000); and Thio et al., “Enhanced Light Transmission through a Single Sub-Wavelength Aperture”, Optics Lett. 26, 1972(2001). They also indicate that a circular concentric pattern is optimal for transmission through a single hole.
Clearly proper application of this surface plasmon could be effective to increase the optical data storage density on phase change optical recording media. What is needed is an effective plasmon lens that can channel sufficient quantities optical energy through a sub-wavelength apertures to produce sub-wavelength spots on optical recording media.